IBM Using Docker and Kubernetes to Enable Watson Cognitive Computing
IBM's Watson computing platform is helping to enable a new era of what IBM calls "Cognitive Computing" with its powerful processing and analytics capabilities. Looking beyond the physical hardware behind IBM Watson, it is Docker containers that are helping IBM to deliver Watson services.
In a video interview with eWEEK, Jason McGee, VP and CTO for IBM cloud platform discussed the intersection of Watson and containers.
"All the Watson services run in Docker containers," McGee said. "So we're running many tens of thousands of containers that are basically the internal implementation for all the Watson APIs the run on (IBM) Bluemix."
From an orchestration perspective, McGee said that IBM has used several different approaches but the strategy is to move everything to Kubernetes. McGee said that Watson makes use of the same container service container capabilities that IBM exposes to its customers. As such, IBM's effort to support containers that help to enable Watson also ends up benefiting IBM's container customers as well that broader open-source container community where IBM contributes code.
"We're right on the cusp of mainstream adoption of containers in production settings," McGee said. "Everyone is using containers in development scenarios and obviously there are lots of people using it in production but they tend to be early adopters. We're now getting into mass adoption on the operational side ."
Watch the full video interview with Jason McGee, VP and CTO for IBM above.
Sean Michael Kerner is a senior editor at eWEEK and InternetNews.com. Follow him on Twitter @TechJournalist.